MFKnockoffs.create.approximate_gaussian: Sample approximate second-order multivariate Gaussian...

In MFKnockoffs: Model-Free Knockoff Filter for Controlled Variable Selection

Description

Samples approximate second-order multivariate Gaussian knockoff variables for the original variables.

Usage

MFKnockoffs.create.approximate_gaussian(X, method = c("asdp", "equi", "sdp"),
  shrink = F)

Arguments

X	normalized n-by-p realization of the design matrix
method	either 'equi', 'sdp' or 'asdp' (default:'asdp') This will be computed according to 'method', if not supplied
shrink	whether to shrink the estimated covariance matrix (default: FALSE)

Details

If the argument shrink is set to TRUE, a James-Stein-type shrinkage estimator for the covariance matrix is used instead of the traditional maximum-likelihood estimate. This option requires the package corpcor. Type ?corpcor::cov.shrink for more details.

Even if the argument shrink is set to FALSE, in the case that the estimated covariance matrix is not positive-definite, this function will apply some shrinkage.

To use SDP knockoffs, you must have a Python installation with CVXPY. For more information, see the vignette on SDP knockoffs: vignette('sdp', package='MFKnockoffs')

Value

n-by-p matrix of knockoff variables

References

Candes et al., Panning for Gold: Model-free Knockoffs for High-dimensional Controlled Variable Selection, arXiv:1610.02351 (2016). https://statweb.stanford.edu/~candes/MF_Knockoffs/index.html

See Also

Other methods for creating knockoffs: MFKnockoffs.create.fixed, MFKnockoffs.create.gaussian

MFKnockoffs documentation built on May 2, 2019, 6:33 a.m.